1 /* $OpenBSD: str.c,v 1.14 2021/11/02 03:09:15 cheloha Exp $ */ 2 /* $NetBSD: str.c,v 1.7 1995/08/31 22:13:47 jtc Exp $ */ 3 4 /*- 5 * Copyright (c) 1991, 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 3. Neither the name of the University nor the names of its contributors 17 * may be used to endorse or promote products derived from this software 18 * without specific prior written permission. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 23 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 30 * SUCH DAMAGE. 31 */ 32 33 #include <sys/types.h> 34 35 #include <assert.h> 36 #include <errno.h> 37 #include <stddef.h> 38 #include <stdio.h> 39 #include <stdlib.h> 40 #include <string.h> 41 #include <ctype.h> 42 #include <err.h> 43 44 #include "extern.h" 45 46 static int backslash(STR *); 47 static int bracket(STR *); 48 static int c_class(const void *, const void *); 49 static void genclass(STR *); 50 static void genequiv(STR *); 51 static int genrange(STR *); 52 static void genseq(STR *); 53 54 int 55 next(s) 56 STR *s; 57 { 58 int ch; 59 60 switch (s->state) { 61 case EOS: 62 return (0); 63 case INFINITE: 64 return (1); 65 case NORMAL: 66 switch (ch = *s->str) { 67 case '\0': 68 s->state = EOS; 69 return (0); 70 case '\\': 71 s->lastch = backslash(s); 72 break; 73 case '[': 74 if (bracket(s)) 75 return (next(s)); 76 /* FALLTHROUGH */ 77 default: 78 ++s->str; 79 s->lastch = ch; 80 break; 81 } 82 83 /* We can start a range at any time. */ 84 if (s->str[0] == '-' && genrange(s)) 85 return (next(s)); 86 return (1); 87 case RANGE: 88 if (s->cnt-- == 0) { 89 s->state = NORMAL; 90 return (next(s)); 91 } 92 ++s->lastch; 93 return (1); 94 case SEQUENCE: 95 if (s->cnt-- == 0) { 96 s->state = NORMAL; 97 return (next(s)); 98 } 99 return (1); 100 case SET: 101 if ((s->lastch = s->set[s->cnt++]) == OOBCH) { 102 s->state = NORMAL; 103 return (next(s)); 104 } 105 return (1); 106 default: 107 return 0; 108 } 109 /* NOTREACHED */ 110 } 111 112 static int 113 bracket(s) 114 STR *s; 115 { 116 char *p; 117 118 switch (s->str[1]) { 119 case ':': /* "[:class:]" */ 120 if ((p = strstr((char *)s->str + 2, ":]")) == NULL) 121 return (0); 122 *p = '\0'; 123 s->str += 2; 124 genclass(s); 125 s->str = (unsigned char *)p + 2; 126 return (1); 127 case '=': /* "[=equiv=]" */ 128 if ((p = strstr((char *)s->str + 2, "=]")) == NULL) 129 return (0); 130 s->str += 2; 131 genequiv(s); 132 return (1); 133 default: /* "[\###*n]" or "[#*n]" */ 134 if ((p = strpbrk((char *)s->str + 2, "*]")) == NULL) 135 return (0); 136 if (p[0] != '*' || strchr(p, ']') == NULL) 137 return (0); 138 s->str += 1; 139 genseq(s); 140 return (1); 141 } 142 /* NOTREACHED */ 143 } 144 145 typedef struct { 146 char *name; 147 int (*func)(int); 148 int *set; 149 } CLASS; 150 151 static CLASS classes[] = { 152 { "alnum", isalnum, }, 153 { "alpha", isalpha, }, 154 { "blank", isblank, }, 155 { "cntrl", iscntrl, }, 156 { "digit", isdigit, }, 157 { "graph", isgraph, }, 158 { "lower", islower, }, 159 { "print", isprint, }, 160 { "punct", ispunct, }, 161 { "space", isspace, }, 162 { "upper", isupper, }, 163 { "xdigit", isxdigit, }, 164 }; 165 166 static void 167 genclass(STR *s) 168 { 169 CLASS *cp, tmp; 170 size_t len; 171 int i; 172 173 tmp.name = (char *)s->str; 174 if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) / 175 sizeof(CLASS), sizeof(CLASS), c_class)) == NULL) 176 errx(1, "unknown class %s", s->str); 177 178 /* 179 * Generate the set of characters in the class if we haven't 180 * already done so. 181 */ 182 if (cp->set == NULL) { 183 cp->set = reallocarray(NULL, NCHARS + 1, sizeof(*cp->set)); 184 if (cp->set == NULL) 185 err(1, NULL); 186 len = 0; 187 for (i = 0; i < NCHARS; i++) { 188 if (cp->func(i)) { 189 cp->set[len] = i; 190 len++; 191 } 192 } 193 cp->set[len] = OOBCH; 194 len++; 195 cp->set = reallocarray(cp->set, len, sizeof(*cp->set)); 196 if (cp->set == NULL) 197 err(1, NULL); 198 } 199 200 s->cnt = 0; 201 s->state = SET; 202 s->set = cp->set; 203 } 204 205 static int 206 c_class(a, b) 207 const void *a, *b; 208 { 209 return (strcmp(((CLASS *)a)->name, ((CLASS *)b)->name)); 210 } 211 212 /* 213 * English doesn't have any equivalence classes, so for now 214 * we just syntax check and grab the character. 215 */ 216 static void 217 genequiv(s) 218 STR *s; 219 { 220 if (*s->str == '\\') { 221 s->equiv[0] = backslash(s); 222 if (*s->str != '=') 223 errx(1, "misplaced equivalence equals sign"); 224 } else { 225 s->equiv[0] = s->str[0]; 226 if (s->str[1] != '=') 227 errx(1, "misplaced equivalence equals sign"); 228 } 229 s->str += 2; 230 s->cnt = 0; 231 s->state = SET; 232 s->set = s->equiv; 233 } 234 235 static int 236 genrange(s) 237 STR *s; 238 { 239 int stopval; 240 unsigned char *savestart; 241 242 savestart = s->str; 243 stopval = *++s->str == '\\' ? backslash(s) : *s->str++; 244 if (stopval < (u_char)s->lastch) { 245 s->str = savestart; 246 return (0); 247 } 248 s->cnt = stopval - s->lastch + 1; 249 s->state = RANGE; 250 --s->lastch; 251 return (1); 252 } 253 254 static void 255 genseq(s) 256 STR *s; 257 { 258 char *ep; 259 260 if (s->which == STRING1) 261 errx(1, "sequences only valid in string2"); 262 263 if (*s->str == '\\') 264 s->lastch = backslash(s); 265 else 266 s->lastch = *s->str++; 267 if (*s->str != '*') 268 errx(1, "misplaced sequence asterisk"); 269 270 switch (*++s->str) { 271 case '\\': 272 s->cnt = backslash(s); 273 break; 274 case ']': 275 s->cnt = 0; 276 ++s->str; 277 break; 278 default: 279 if (isdigit(*s->str)) { 280 s->cnt = strtol((char *)s->str, &ep, 0); 281 if (*ep == ']') { 282 s->str = (unsigned char *)ep + 1; 283 break; 284 } 285 } 286 errx(1, "illegal sequence count"); 287 /* NOTREACHED */ 288 } 289 290 s->state = s->cnt ? SEQUENCE : INFINITE; 291 } 292 293 /* 294 * Translate \??? into a character. Up to 3 octal digits, if no digits either 295 * an escape code or a literal character. 296 */ 297 static int 298 backslash(STR *s) 299 { 300 size_t i; 301 int ch, val; 302 303 assert(*s->str == '\\'); 304 s->str++; 305 306 /* Empty escapes become plain backslashes. */ 307 if (*s->str == '\0') { 308 s->state = EOS; 309 return ('\\'); 310 } 311 312 val = 0; 313 for (i = 0; i < 3; i++) { 314 if (s->str[i] < '0' || '7' < s->str[i]) 315 break; 316 val = val * 8 + s->str[i] - '0'; 317 } 318 if (i > 0) { 319 if (val > UCHAR_MAX) 320 errx(1, "octal value out of range: %d", val); 321 s->str += i; 322 return (val); 323 } 324 325 ch = *s->str++; 326 switch (ch) { 327 case 'a': /* escape characters */ 328 return ('\7'); 329 case 'b': 330 return ('\b'); 331 case 'f': 332 return ('\f'); 333 case 'n': 334 return ('\n'); 335 case 'r': 336 return ('\r'); 337 case 't': 338 return ('\t'); 339 case 'v': 340 return ('\13'); 341 default: /* \x" -> x */ 342 return (ch); 343 } 344 } 345